Abstract Strong evidence from animal studies and indirect data from human imaging studies, implicate reduced liver mitochondrial activity as a key event in the pathophysiology of nonalcoholic steatohepatitis (NASH). Our preliminary data in NASH patients demonstrate that two treatments that relieve excess nutrient stress, caloric restriction (CR) and exercise training (EX), improve liver histology. Our recent studies in a rodent model established that the therapeutic effect of EX on liver histology is mediated through increases in both hepatic mitochondrial content and function. It is unknown whether a similar beneficial effect of such treatment on liver mitochondria occurs in humans. The central hypotheses of this project is that nutrient overload results in hepatic mitochondrial dysfunction, a key mediator of NASH pathology, and that amelioration of this overload will result in greater efficiency of mitochondrial bioenergetics and improved liver histology. To test these hypotheses, a treatment regimen combining CR and EX will be used to reduce intrahepatic triacylglycerols (IHTG) in 60 NASH patients. Baseline and 9-month liver biopsies in the treated group will be compared to repeat biopsies from 30 patients undergoing standard clinical management. We will complete the following specific aims. SA1 will utilize liver biopsy samples from patients with NASH, and liver samples obtained during bariatric surgery, to determine whether patients with more advanced liver disease have liver mitochondria that, in vitro and ex vivo, exhibit poor energy generation (quality control), biogenesis, and mitophagy; SA2 will isolate hepatic mitochondria again after 9-months to test whether loss of IHTG is associated with improved mitochondrial function, biogenesis and quality control as assessed by gene expression, enzyme levels and fatty acid oxidation. SA3 will measure whole body glucose and fatty acid flux in subjects under- going active treatment (CR+EX) and standard care. Measures will be made at baseline and after 9 months, and mathematical modeling used to quantitate peripheral nutrient disposal and the reduction in liver nutrient burden to compare its influence on mitochondrial function, fibrosis, and by the biopsy- determined NASH activity score using histological assessment. In summary, these studies will significantly advance the field of NAFLD through discoveries of 1) the mechanisms by which excess nutrient stress contributes to reduced hepatic mitochondrial function and 2) the relationships between mitochondrial activity and the histologic features of NASH. Importantly, this work will exert a sustained influence on the field by determining whether hepatic mitochondrial dysfunction is reversible in humans.